Cancer prognosis

Abstract

A computer-implemented for parameterizing a statistical function. The statistical function is configured to perform prognosis on cancer patients. The parameterizing method includes obtaining a dataset relative to a plurality of patients each having a cancer disease. The dataset includes, for each patient, input data of the statistical function. The input data include visual data from at least one histological image of a tumor slide of the patient. The input data also include genomic data from sequencing tumor tissue of the patient. The input data further include clinical data of the patient. The dataset also includes clinical endpoints of the patient with respect to evolution of the cancer disease. The parameterizing method further includes parameterizing the statistical function based on the dataset. This forms an improved solution to perform prognosis on patients having a cancer disease.

1 . A computer-implemented method for parameterizing a statistical function configured for performing prognosis on cancer patients, the method comprising:
 obtaining a dataset relative to a plurality of patients each having a cancer disease, the dataset including for each patient:   input data of the statistical function, the input data including:
 visual data from at least one histological image of a tumor slide of the patient, 
 genomic data from sequencing tumor tissue of the patient, and 
 clinical data of the patient, and clinical endpoints of the patient with respect to evolution of the cancer disease; and 
   parameterizing the statistical function based on the dataset.   
     
     
         2 . The computer-implemented method of  claim 1 , wherein the visual data are obtained from one or more maps representing one or more distributions including:
 a distribution of tissue textures,   a distribution of lymphocyte presence, and/or   a distribution of cell density.   
     
     
         3 . The computer-implemented method of  claim 2 , wherein at least one map is extracted from the at least one histological image via a respective convolutional neural network (CNN). 
     
     
         4 . The computer-implemented method of  claim 2 , wherein the one or more maps include several maps each representing a respective distribution. 
     
     
         5 . The computer-implemented method of  claim 3 , wherein:
 the one or more maps comprise a texture map extracted from the at least one histological image via a first CNN, the texture map having pixels each representing a texture among a predetermined set of textures, each pixel resulting from classifying a respective tile of the at least one histological image with the first CNN,   the one or more maps comprise a lymphocyte map extracted from the at least one histological image via a second CNN, the lymphocyte map having pixels each indicative of lymphocyte presence, each pixel resulting from classifying a respective tile of the at least one histological image with the second CNN, and/or   the one or more maps comprise a cell density map extracted from the at least one histological image via a third CNN, the cell density map having pixels each representing a local number of cells, each pixel resulting from counting a number of cells in a respective tile of the at least one histological image with the third CNN.   
     
     
         6 . The computer-implemented method of  claim 1 , wherein the genomic data represent a cell-type distribution and/or gene expression data. 
     
     
         7 . The computer-implemented method of  claim 1 , wherein the clinical data represent at least one among patient age, patient gender, tumor stage, lymph node stage and/or metastasis stage. 
     
     
         8 . The computer-implemented method of  claim 1 , wherein the statistical function comprises a neural network, a stochastic model, and/or a regression model such as a Cox model. 
     
     
         9 . The computer-implemented method of  claim 1 , wherein the statistical function is configured to output a prediction on patient's outcome and/or a prediction on tumor evolution. 
     
     
         10 . A computer-implemented method of implementing a statistical function for performing prognosis on a cancer patient, the statistical function being parameterized by parameterizing a statistical function configured to perform prognosis on cancer patients, the method comprising:
 obtaining input data including: visual data from at least one histological image of a tumor slide of the patient, genomic data from sequencing tumor tissue of the patient, and clinical data of the patient; and   applying the statistical function to the input data to perform prognosis on the patient,   wherein the parameterizing the statistical function includes:   obtaining a dataset relative to a plurality of patients each having a cancer disease, the dataset including for each patient:
 input data of the statistical function, the input data including: visual data from at least one histological image of a tumor slide of the patient, genomic data from sequencing tumor tissue of the patient, and clinical data of the patient, and 
 clinical endpoints of the patient with respect to evolution of the cancer disease; and 
   parameterizing the statistical function based on the dataset   
     
     
         11 . A device comprising:
 a non-transitory computer-readable data storage medium having recorded thereon a computer program comprising instructions for parameterizing a statistical function configured for performing prognosis on cancer patients that when executed by a processor causes the processor to be configured to:   obtain a dataset relative to a plurality of patients each having a cancer disease, the dataset including for each patient:
 input data of the statistical function, the input data including: visual data from at least one histological image of a tumor slide of the patient, genomic data from sequencing tumor tissue of the patient, and clinical data of the patient, and 
 clinical endpoints of the patient with respect to evolution of the cancer disease; and 
   parameterize the statistical function based on the dataset.   
     
     
         12 . The device of  claim 11 , wherein the visual data are obtained from one or more maps representing one or more distributions including:
 a distribution of tissue textures,   a distribution of lymphocyte presence, and/or   a distribution of cell density.   
     
     
         13 . The device of  claim 12 , wherein at least one map is extracted from the at least one histological image via a respective convolutional neural network (CNN). 
     
     
         14 . The device of  claim 12 , wherein the one or more maps include several maps each representing a respective distribution. 
     
     
         15 . The device of  claim 14 , wherein:
 the one or more maps comprise a texture map extracted from the at least one histological image via a first CNN, the texture map having pixels each representing a texture among a predetermined set of textures, each pixel resulting from classifying a respective tile of the at least one histological image with the first CNN,   the one or more maps comprise a lymphocyte map extracted from the at least one histological image via a second CNN, the lymphocyte map having pixels each indicative of lymphocyte presence, each pixel resulting from classifying a respective tile of the at least one histological image with the second CNN, and/or   the one or more maps comprise a cell density map extracted from the at least one histological image via a third CNN, the cell density map having pixels each representing a local number of cells, each pixel resulting from counting a number of cells in a respective tile of the at least one histological image with the third CNN.   
     
     
         16 . The device of  claim 11 , wherein the device further comprises the processor coupled to the non-transitory computer-readable data storage medium. 
     
     
         17 . The device of  claim 12 , wherein the device further comprises the processor coupled to the non-transitory computer-readable data storage medium. 
     
     
         18 . The device of  claim 13 , wherein the device further comprises the processor coupled to the non-transitory computer-readable data storage medium. 
     
     
         19 . A non-transitory computer readable medium having stored thereon a program that when executed by a computer causes the computer to implement the method for parameterizing the statistical function configured for performing prognosis on cancer patients according to  claim 1 . 
     
     
         20 . A non-transitory computer readable medium having stored thereon a program that when executed by a computer causes the computer to implement the method for implementing the statistical function for performing prognosis on the cancer patient according to  claim 10 .